In a study at Children’s Hospital, Ann Arbor, MI, and Children’s Memorial Hospital, Chicago, designed primarily to compare cost-effectiveness of three methods of management of cyclic vomiting cases, one group of patients received an extensive diagnostic evaluation, a second was treated with empiric antimigraine drugs for 2 months, and a third an upper GI series with small-bowel follow-through (UGI-SBFT) plus empiric therapy. Cyclic vomiting is defined as >3 episodes of vomiting within a 3 month period, peak intensity averaging 6 emeses per hour, and intervals of normal health averaging 2 to 4 weeks. The most cost-effective approach was the UGI-SBFT followed by antimigraine therapy. A CT scan before antimigraine therapy to rule out a brain tumor would be cost-saving, even though the prevalence of brain tumor was 0.5% (range 0-2%). Except for a missed diagnosis of volvulus resulting in extensive small-bowel resection, the cost of delay in diagnosis of other causes of cyclic vomiting (eg metabolic) were less than the diagnostic tests. [1]

COMMENT. According to these authors and based on cost-effectiveness, the most practical approach to the diagnosis of cyclic vomiting(CV) is an initial small-bowel radiograph to rule out malrotation followed by a 2-month empiric trial of antimigraine medication. A positive family history of migraine is reported in 82% of patients with idiopathic CV, and their response to migraine prophylaxis is better than those without a family history of migraine (79% vs 36%) [2]. Only 12% have serious underlying disorders such as intestinal malrotation with volvulus, brain tumors, or metabolic disorders.

Cyclic vomiting is reported as a form of epilepsy in children [3], and ictus emeticus with nondominant temporal lobe involvement is a well documented form of autonomic epilepsy, sometimes induced by photic stimulation [4]. See Progress in Pediatric Neurology I & III. 1991 & 1997, for further review of ictus emeticus.

Vomiting as an ictal phenomenon is controversial and difficult to distinguish from migraine. Symptoms should be paroxysmal and associated with ictal epileptiform discharges on the EEG. In our own retrospective study of 33 children with cyclic vomiting reported in 1955, 25 (76%) had interictal seizure discharges in the EEG, some focal with temporal localization, and 7 (21%) had a history of complex partial or generalized seizures. Epilepsy should certainly be considered in the differential diagnosis of cyclic vomiting, an omission in the above report, and an EEG obtained, if possible during a vomiting episode. A history of previous seizures and family history of epilepsy, brain pathology, and a beneficial response to antiepileptic drugs will help to corroborate the diagnosis. An abnormal EEG is common in children with migraine, and AEDs are an effective migraine prophylaxis [5], a further confounding factor in the diagnosis of cyclic vomiting and differentiation of an epilepsy or migraine.